Becoming Attached: First Relationships and How They Shape Our Capacity for Love

Home , Mary Ainsworth, Mistress (lover)

Karen, Robert. Becoming Attached: First Relationships and How They Shape Our Capacity for Love. New York: Oxford University Press, 1998.

Robert Karen works as a private practice clinical psychologist and associate clinical professor at the Derner Institute of Advanced Psychology Studies of Adelphi University.

Introduction: How do we become who we are? Everyone seeks love. Many encounter difficulties in achieving satisfying relationships. Parents set the stage for the emotional tenor of children. Attachment theory addresses the centrality of love between humans, how its roots emerge in parent-child relationships, how that relationship forms and endures, and the effects of damage to or insufficiency in the relationship. Attachment theory is deeply debated. Eugenics theorists argue children become what their genes dictate. Behaviorists argue that children are infinitely malleable. Psychoanalysts moderated these extremes. John Bowlby shaped attachment theory, arguing that first relationships determine much about future affection. Mary Ainsworth followed up Bowlby’s theories with field research on mothers and babies in their homes. She found secure attachment (welcoming, responsive, attuned, and reliable) created emotional health. One-third of American children suffer insecure attachment. Attachment, secure or insecure, influences all of life. It echoes forward through time, affecting even a person’s offspring. Insecure attachments can change, with some effort. Some attachment theorists view day care as a threat to child well-being, because the child may become anxiously attached. Feminists have objected to the over-emphasis on maternal influence. Other critics argue that attachment theorists under-appreciate the role of heredity in child and adult behavior. This book considers attachment, the evidence for attachment, and the arguments of the critics of attachment.

Part I. What Do Children Need?

1. Mother-Love: Worst-Case Scenarios. Lacking a mother may cause inability to form relationships or interact with the world with normal emotionality. Mother-love suffers many injuries, any of which may lead to a mother’s inability to care for her baby’s difficulties. If mother’s love is entirely absent, dire consequences follow for children. David Levy studied such circumstances. Children deprived in this way become shallow in affection, indifferent, absent pride of self, unmanageable (including sexual aggression, lying, theft, outbursts), and lacking in friendships. Levy postulated emotional starvation. Loretta Bender’s research concurred. The children she investigated suffered emotional deprivation during formative years. In hospitals, concern for sterility led to a rule that none handled babies. Untouched babies failed to thrive and died in droves. When the rule was inverted, and all were urged to hold babies, survival rates soared. Inadequate attention to infant emotional needs damaged intelligence. A potent documentary on infant isolation (Grief: A Peril in Infancy) by Rene Spitz shocked child workers. It showed healthy babies collapsing into grief when mothers left them in good care for long periods. Some babies suffered permanent injury or even died.

2. Enter Bowlby: The Search for a Theory of Relatedness. John Bowlby’s work drew further attention to infant-mother relations. The emotional quality of the home governs. Damage ensues if the infant is separated from its mother for a long period. Damage also ensues if the mother’s emotions toward the infant are ambivalent. Personally, Bowlby was affable, but arrogant, and occasionally aggressive with unwanted interruption or dissent. Yet, his good manners helped him keep up relations with opponents. In youth, Bowlby suffered typical British upper-class distance from his parents, and boarding school at the age of eight. As an adult, Bowlby was taken by a fringe view of child-rearing: let them do as they please, and support their learning. Teaching at such a school, Bowlby saw the connection between troubled homes and disturbed behavior in children. Bowlby studied psychoanalysis. He resisted his own treatment. He joined Canonbury Clinic, a family mental health center. There, other workers showed him that a mother’s unresolved emotional issues affected her parenting of her infant. They began treating both mother and child. Freud argued that neurosis found its source in childhood sexual trauma, real or imagined (Oedipal complex). Freudian analysts neglected environmental influences, favoring

psychological ones. They neglected the real relationships their patients had with parents. Non- Freudians began arguing that humans have a fundamental need for meaningful connection to others, as deep a need as food or sex. Fairbairn criticized Freudian analysis, arguing that psychopathology emerged from early relational disturbances.

3. Bowlby and Klein: Fantasy vs. Reality. Melanie Klein developed psychoanalytic play theory. She argued that the maternal relationship becomes a template for adult relationships in every person. Klein found that the infant’s imagined mother is more real than the actual mother, and the impact of the fantasy mother on the infant may be poorly correlated with the actual mother’s mothering. Infants do not see persons, but pieces of persons: faces, breasts, hands. After six months, whole persons become present for the infant. For infants, the breast is critical, according to Klein. It is a good or evil breast, according to the experience of the infant with it. The infant rages, though not intellectually. Infants have a violent inner life. All future experience repeats this early drama of the good and evil breast. Klein described the “depressive position” that emerges for the child when it recognizes that the evil and good breasts are in fact the one person of the mother. Klein’s views were ill-received by many. Klein’s own difficult personality (devious and nasty) and ill-considered ethical choices (analyzing her own daughter) compounded the vigor of dissent to her views. Melanie Klein became John Bowlby’s supervisor. Klein immediately refused to let Bowlby treat disturbed children’s social setting. Only the child’s internalization mattered. Bowlby revolted, seeing clearly that, in his own view, real life experiences were the crux, not something to be regarded as unimportant. World War II intervened. Bowlby served, and Klein’s supervision ended. Klein influenced Bowlby to consider infantile fantasy in his clinical and theoretical work.

4. Psychopaths in the Making: Forty-four Juvenile Thieves. Bowlby believed that parental response to children’s inner conflicts diverted children down a path toward coping with their dramatic emotions or being controlled by them. Defenses bury unwanted feelings beneath consciousness: humor, sublimation, repression, reaction formation, passive-aggression, acting out, delusional projection, and reality denial. Bowlby researched forty-four adolescent thieves at the Child Guidance Center. All had poor or no parental relating in early life. Upon examination, all the parents were very dysfunctional. Bowlby needed a group of normal children and families to compare and contrast. But that lay beyond his resources and so a longitudinal study of these observations could not be mounted. So Bowlby turned his attention to one factor only: early prolonged separation from mother. Among his thieves who were affectionless (psychopathic), twelve of fourteen suffered early maternal disruption. Bowlby argued that symbolically the thefts were attempts to gather the affection missing from early life. The dearth of early affection also worked itself out in adult sexuality. Deprived children imagine their mothers as hateful, and so end up hating themselves. Interrupting this vicious circle perhaps lay beyond the coping skill of most parents. Bowlby’s paper on the forty-four thieves was influential. Later research showed that mere separation did not cause psychopathy, but rather unremediated separation, where the child has no opportunity to establish true attachment to anyone.

5. Call to Arms: The World Health Report. The United Nations conducted a study of the needs of homeless children after World War II. Bowlby wrote a report on this subject, Maternal Care and Mental Health, which amalgamated research from Europe and the United States. Bowlby argued that maternal care was essential to infants, and separations injure both infants and later society as well. Institutions cannot provide the maternal care that babies elicit naturally from mothers. Neglected children grow up to become neglectful parents, and are a source of “social infection.” Bowlby’s report had massive impact. It generated massive response, both advocates and detractors.

6. First Battlefield: “A Two-Year-Old Goes to Hospital.” In the 1950s and before, hospitalized young children’s parents were excluded. These children suffered emotional trauma. Unable to understand the abandonment, young children despair, and become locked into permanent suspicion and alternatives to attachment (such as aggression, jealousy, theft). All suffered

permanent insecurity about maternal abandonment. Even short separations (mere days) can cause deep emotional injury. Bowlby took over the Tavistock Clinic children’s department. His associate, James Robertson, and Robertson’s wife, Joyce, chronicled the emotional problems of young children separated from parents in the hospital ward. Robertsons’ research was disregarded by hospital staff. Robertsons made a film of Laura, a two and one-half year old in hospital for eight days. The effects on self-possessed Laura were profound and enduring. Robertsons were savaged by hospital staffs for slandering them. But reviews of the research were generally positive, and deeply influential. A Glasgow hospital allowed one children’s ward unrestricted maternal visits, and another would continue to follow the one-half hour twice weekly visitation schedule for parents. The experiment became the rule for the hospital. Robertsons made a follow-up film on going to hospital with mother. The attended child suffered none of the ill-effects of maternal absence. A series of subsequent films followed healthy young children separated from their ill mothers under nearly ideal conditions. All suffered separation anxiety that changed the course of their development somewhat. The one film Robertsons showed a seventeen-month old child in institutional care while his mother had another baby. The seventeen-month old suffered the expected damage, and it persisted for months after the nine day separation. Robertsons became alienated from Bowlby, though Bowlby’s appreciation of the Robertsons persisted.

7. Of Goslings and Babies: The Birth of Attachment Theory. Bowlby encountered the imprinting work of Konrad Lorenz. Goslings imprint upon the first moving object they see after hatching. Ethologists (ethology studies animal behavior in natural settings), starting with Darwin, asserted that each species exhibits specific behaviors, as unique as their anatomy. Bowlby applied this learning to human infants, asserting humans are hard-wired for relationship, starting with mother. A baby’s need for an attentive and reliable mother-figure is instinctive, just as providing that attention is instinctive for parents. Bowlby coined the term “attachment” for the constellation of behaviors a baby undertakes to elicit response from his caregivers (cooing, sucking, babbling, smiling, crying at departures, welcoming returns, clinging, following), and the responsive of the caregivers themselves to that infant. Attachment, in Bowlby’s characterization, lies close to love. Having it, an infant prospers. Lacking attachment at critical junctures, depression, anxiety, and suffering follow for an infant. By 1950, most psychologists agreed that a child’s earliest relationships are foundational and that mother is preeminent among those. But no one had a convincing theory. Buhler worked on the early facial “talk” between smiling mother and newborns. Balint argued that humans wish to be loved without conditions. If they lack this in childhood, they spend the remainder of life hunting for it. Winnicott’s theory most strongly resembled Bowlby’s, but they were very different sorts of people, and collaborated poorly. Bowlby delivered three papers. He argued that babies’ behaviors aim to elicit care and stimulate parents to provide that care. It is multi-form, and not based solely on feeding. Neurosis derives from disturbances to a child’s attachment system. By six or seven months of age, the infant recognizes mother in particular, and rejects parenting by other competent replacements. (The author, Robert Karen, disputes Bowlby on this count, arguing anecdotally that babies even earlier in their development show preference for their parents.) Bowlby called the child’s hierarchy of care-giving preference “monotropy.” Maternal absence produces astonishing anxiety in the abandoned child. Bowlby believe that attachment-disrupted children grieve as do adults suffering loss. Normal grief resolves. Pathological grief fails to resolve. Attachment disrupted children may never resolve their grief. As Freud developed an entire psychological framework commencing from hysteria, Bowlby started with separation.

8. “What’s the Use to Psychoanalyze a Goose?” Turmoil, Hostility, and Debate. Some mothers, reading Bowlby’s and Robertsons’ works, came to dread leaving their babies for even a few hours. Bowlby sought to comfort and reassure these parents. Bowlby sympathized with the pressing emotional and physical demands of having a young child, and warned mothers that when they have babies, they face five years of hard labor. If a substitute caregiver (say, a nanny) spends a great deal of time with the child, she must remain until the child reaches six years of age. She will become the emotional “mother.” Bowlby recommended against child hospitalizations, and insisted mothers stay with children who must be hospitalized. When separation was

unavoidable, Bowlby recommended the child be cared for by a familiar person who is willing to provide care full-time during mother’s absence. Even with such care, the child will be disturbed by mother’s absence. Re-attachment will require time and great affection and reassurance. Bowlby created an ancillary furor by asserting that father’s parental role was to support mother. His absence at work precluded him providing fundamental security to his children. Bowlby urged mothers never to stint on their giving their child attention and affection, for, if absent, one can never really replace what was lost. Over time, Bowlby came to recognize that humans better make up early losses than do goslings. Still, he recommended that mothers take a leave of absence until their young children mature (age 3-6). Critics (Melanie Klein and Anna Freud) emphasized that a child’s fantasies about maternal inattention were of as much effect as real inattention. Bowlby countered that fantasies find their first brick in reality. Freudian critics argued that Bowlby so focused on human relations as to neglect what was going on in the child’s head. Babies are not fish or birds. What distinguishes them is what goes on in babies’ heads. Anna Freud and Melanie Klein were dogmatic authoritarian personalities. All who disagreed with them, were, in their view, wrong. Bowlby disagreed, and held all psychoanalytical pronouncements up to experimental scrutiny. Bowlby himself could be utopian, imagining that once parenting disasters were avoided, man would be less tortured and more amicable. Some colleagues believed Bowlby aimed to destroy psychoanalysis. For many analysts, allegiance to Freud’s theory was inseparable from their work and identities. Bowlby’s theory could not be adapted (or Bowlby would not make the effort to adapt it). The British psychoanalytic conventions at which Bowlby read his papers roiled. Sessions were added to accommodate the extended debates. Bowlby withdrew from the Psychoanalytic Society, so threatened by his work, and interacted little for the next three decades. Some research undercut Bowlby’s previous research. Children responded differently to maternal deprivation. Character deformation proved more complex than mere social disruption. Critics tore apart many studies’ methodologies or reporting standards. The case for Bowlby’s views sagged.

9. Monkey Love: Warm, Secure, Continuous. Harry Harlow at University of Wisconsin had disease-borne deaths among his experimental rhesus monkeys. He separated sixty babies shortly after birth to an antiseptic isolation. The babies bonded to the cloth diapers on the cage floors. Subsequent experiments reproduced Bowlby’s child observations in rhesus infants. Cuddling mattered more than breasts. All attached to their woefully defective wire mesh mother- surrogates. All suffered later difficulty relating or raising their own young. Bowlby went to Wisconsin to see Harlow; they remained in touch with one another’s work. The United Nations elected to do a follow-up study to Bowlby’s 1951 report. Mary Ainsworth drafted the core review. She took apart criticisms of Bowlby’s work, and sorted maternal deprivations into three types: insufficiency, neglect, and separation. This resolved many internal conflicts in Bowlby’s work. The rhesus monkey work continued, showing results that supported Bowlby’s theories. Ainsworth concluded that Bowlby’s work was essentially sound.

Part II. Breakthrough: The Assessment of Parenting Style

10. Ainsworth in Uganda. Ainsworth, already a faculty member, followed her younger husband to London, where she answered a help-wanted ad placed by Bowlby. They worked together for more than three years. Later, Ainsworth studied the mother-child relationship in the child’s first year in their natural environment in Uganda. Ainsworth observed twenty-eight babies in their familial settings for a period of nine months, visiting each for about two hours every other week. Ainsworth noted the milestones in infant development and attachment in detail, concluding that mother provided a secure base from which a child can, without anxiety, move out to explore her environment. Ainsworth found five phases of attachment development: a) undiscriminating (newborn), b) differential responsiveness, in which the baby shows preference for the mother, c) differential preference at a distance, in which the baby greets or objects to the mother’s comings and goings, d) active initiative, in which the infant follows the mother and begins exploring around the mother, and e) stranger anxiety, in which the infant recognizes outsiders and watches them. Parenting style mattered in attachment. Infants failed to attach to distant mothers, but attached to others who were responsive. Five of her twenty-eight (18%) failed to attach.

Ainsworth struggled to differentiate why some babies did not cry when mothers left them; was it a secure relationship or failure of attachment? Seven babies were insecurely attached. These babies initially looked secure and attached, but grew less so as the months passed. Their mothers were dutiful and affectionate, but distracted. Total amount of care provided by extended family did not prove significant. The securely attached infants got the most care, the unattached the least, and the insecurely attached were in between. Ainsworth’s book, Infancy in Uganda, established the concept of the “secure base.”

11. The Strange Situation. Ainsworth moved to Johns Hopkins University and divorced. Bowlby learned of her Uganda research and saw its relevance to his work. They became intellectual partners. Ainsworth did a longitudinal study of Baltimore mothers and infants. She chose twenty-six families with babies coming. Her team would make eighteen visits of four hours each over the infant’s first year. Ainsworth found all sixteen attachment behaviors the same in Ugandan and American babies (with two cultural deviations). American babies were less insecure than Ugandan about the maternal secure base when at the mother’s home. Ainsworth devised a lab experiment to test whether the secure base idea pertained as in Uganda. A lab with three chairs and toys was established. An infant came with its mother to the lab for an eight episode event. A stranger joins them. The mother leaves. The stranger leaves. The stranger returns. The mother returns. This arrangement was called the “strange situation.” Some insecurely attached babies experienced great difficulty with the separations, and only slowly reestablished connection with the mother, all the while both wanting and resisting reconnection (ambivalent babies). Others were barely perturbed by the separations, and sought less comfort from the mother. These often expressed some hostility or even anger (avoidant babies). Ainsworth developed a mothering scale: 1) sensitive to baby’s signals, 2) accepting of the baby, 3) cooperating with baby’s desires, and 4) availability to baby. Securely attached babies have mothers high on these standards. Insecurely attached babies had mothers who scored poorly on these measures. The quality, not quantity, of mothering mattered most. Ainsworth subdivided her three groups (secure, ambivalent, and avoidant) into eight subcategories. These categories became a metric for parenting and attachment. They demonstrated what Bowlby had urged: early parenting behaviors were crucial to a child’s emotional development. And Ainsworth’s work built bridges, much in decay under Bowlby, back to the psychoanalytic community.

12. Second Front: Ainsworth’s American Revolution. Young scholars began adopting attachment theory as their base of experiment. These researchers recognized in Ainsworth’s approach a shift from counting behaviors to exploring the meaning of a child’s and mother’s interactions. The unit of study was no longer the individual, but rather the dyad. These young researchers found that attachment styles persisted into the baby’s future. Ainsworth and her colleagues taught in institutions with behaviorist orientations. Watson and Skinner were pre-eminent. Consciousness was discounted, behavior lionized. Thought was subvocal speech. All human behavior derived from environmental stimuli; none was inherent. Behaviorists had little patience for Ainsworth’s emphasis on relationships and qualitative analysis. And they rejected Bowlby’s view that attachment is an inborn need of and structure within babies. The behaviorist faculties declined to fund Ainsworth’s ongoing studies. But slowly, as the children evaluated in the Strange Situation labs developed, the attachment styles were correlated with a vast host of behaviors in older children. Ainsworth slowly prevailed.

Part III. The Fate of Early Attachments

13. The Minnesota Studies: Parenting Style and Personality Development. Bowlby used the Ainsworth data to argue that infants who enjoyed sensitive responsiveness to their needs developed secure attachment to a trusted figure, which served, as the infant matured, to make her self-reliant, while still trusting in and collaborating with others. Sroufe conducted longitudinal studies of children with known attachment histories to see what came of secure and insecure attachments. Sroufe believed that secure attachment led to high quality adaptation to the changes and new circumstances maturation brings infants. Securely attached children, faced with a difficult play problem, persisted, better managed frustration, appropriately sought assistance, and

sought autonomy. Ambivalent or avoidantly attached children performed less well. Mahler’s research showed that toddlers want mother’s participation in all they do, but also push mother away for growing independence. Mothers of securely attached toddlers remained deeply involved, but now encouraged problem-solving and independence. They gave just what was needed to keep the toddler exploring and solving. Mothers of anxiously or avoidantly attached toddlers failed to keep their distance or stay involved. At three and one-half years, securely attached children were more advanced in other relationships. Egeland conducted the next set of studies, which involved 267 low-income single pregnant women. Egeland studied child abuse; Sroufe studied attachment patterns. Evaluations of mother’s emotional state and enthusiasm for pregnancy predicted the sort of attachment their babies would achieve. One neurological study correlated a predictor with ambivalent attachment (which created a firestorm in later debates). When the children reach age four and one-half, the researchers created a pre-school attended by forty of the sample children. Securely attached children outperformed others in every category of analysis. Secure children were more likely to show empathy. Insecurely or avoidantly attached children were, respectively, more likely to feel personally injured or revel in others’ pain. Sroufe identified three avoidant types: lying bullies, shy detached emotionally flat children, and detached twitching daydreamers. Sroufe identified two ambivalent types: impulsive fidgeters with low concentration, and fearful, sensitive, hiding child with low initiative. All seemed to suffer painful feelings of separation; all were highly dependent. Avoidant children, in their lying and aggression and projected false self-confidence, angered teachers. Ambivalent children, who showed insecurity, elicited sympathy and were judged according to lower standards. The Minnesota studies show the influence of early attachment patterns to emotional patterns of adulthood. The studies, however, failed to prove that secure attachment caused emotional health (temperament must also be a factor), and not all children observed followed the predicted patterns at all. The Minnesota studies faced political opposition, coming at a time when women were entering the work force and revising their parenting styles. Attachment theory exposed that western reliance upon mothers as the generally sole instrument of nurturing young children, with paternal absence, is unworkable, given the fragility of life stability in the first place. We rely on mother nurture more than other cultures and more than other times in history. Attachment theory identified mothers as the problem. Critics seized on this.

14. The Mother, the Father, and the Outside World: Attachment Quality and Childhood Relationships. As the Minnesota subjects aged into social relations with outsiders, the patterns continued. Securely attached children made sensitive partners, formed friendships, and played creatively. Avoidantly attached children did not want relationships and took advantage of weaker playmates. Ambivalently attached children wanted relationships but proved incompetent in forming them. When avoidant children were matched with ambivalent ones, avoidant cruelty often followed. Securely attached children avoid aggression, withdraw or transform relationships that pose dangers, and use force only as a last resort, and then as little as possible. Empathic, peacemaking children have experienced empathic, peaceful core relationships and export that model to their social environment. In early adolescence, social groups formed from entirely securely attached children. Ambivalent children could not navigate the social requirements of participation without alienating their peers. Among the children, sexual differences emerged. Avoidant boys tended toward violence, lying, and boasting. Avoidant girls tended to internalize their stresses, becoming depressed or feeling shame. Sexual differentiation in behavior was minimal in securely attached children. Insecure girls tend to smile inordinately. Ainsworth believed that the system involved in social relations is more complex than early attachment can explain. Sroufe’s longitudinal studies of these Minnesota children may prove that early attachment profiles project far into adult life, affecting even the attachment patterns of children born to the Minnesota subjects. Researchers began to investigate paternal attachment. Distressed infants prefer mothers, but reach out to fathers whenever mothers are absent and in the same ways they reach out to mothers. Among securely attached children, those most secure and empathetic had secure attachment to both parents. The secondary caregiver, usually father, is not a lesser mother. Fathers, when attached to toddlers, provide exciting play and a bridge to the world beyond mother. Fathers offer something mothers cannot. Fathers provide to sons a model with which to

identify. Though fathers often become significantly involved with their children only when they are older, their impact, especially in their satisfaction or dissatisfaction with the mother, proves formative to the partners’ children. These powerful observations about the predictive impact of early attachment patterns created a tidal wave in academic psychology.

15. Structures of the Mind: Building a Model of Human Connection. Why are emotional habits acquired early in life so persistent and predictive of future outcomes? Early emotional experience becomes a blueprint for future events and relationships. Psychoanalysis dealt with these issues by talk of “transference,” but Bowlby sought to clear away the brambles with his “internal working model.” Piaget described an infant’s flexible learning programs, which, elaborated by experience, develop into adult intelligence. Working from Piaget’s research, Bowlby described the internal working model, a mental structure in infants that makes of early relational experience a template for future intimacies, how one views and feels about himself and others. Attachment creates sufficient security to enable exploration. Injured or scared during exploration, an infant returns to the safe haven of the attached adult to recover and refuel. Anxiously attached children have rigid internal working models; their experiences do not update their model as happens in securely- attached children. The anxiously-attached (ambivalent or avoidant) person gives little, distrusts, ignores or misinterprets persons offering friendship, affects superiority, stands off and backs away. Having done so, his future relationships reflect these attitudes, confirming anxious attachment’s predictions. In children, the internal working model is complicated by a large number of distortions and misinterpretations the children suffer, due to their lack of experience. These distorted internal working models may be revised later in life. But the feelings these distortions engender are often repressed, becoming buried and denied. Anxious attachment tends to rob children of the capabilities necessary to work through distorted internal models. Anxiously attached children have reduced ability for self-reflection, less ability to imagine what others might be feeling when it differs from their own feelings, and reduced ability to recall the past accurately. Bowlby’s “internal working model” is now viewed as simplistic, failing to take seriously the depth of misery that dysfunctional psychological patterns may inflict on people. But Bowlby’s work encouraged cross-disciplinary studies and a vast body of empirical research.

16. The Black Box Reopened: Mary Main’s Berkeley Studies. Mary Main, who studied under Ainsworth, collaborated in a longitudinal study of forty six-year olds. By showing them pictures of children separating from parents, the group accurately identified attachment style in seventy- nine percent. Securely attached children were able to discuss what the pictured children might be experiencing. Avoidant children, however, struggled to express themselves, and saw few options for coping. They were flooded with anxiety. Ambivalent children expressed contradictory impulses of wanting proximity and rage at abandonment. They often imagined killing a parent. Disturbed children (a fourth category) expressed intense fear in language that was frequently nonsense. Main’s contribution lay in her recognition that attachment styles affect more than behavior. Attachment determines the structure of a child’s brain, and affects attention, memory, and thinking. Behaviorists had discredited speaking of the mind because it is a black box from which behavior emerges. All mind talk was gibberish to behaviorists. Main disagreed. She set a student, Jude Cassidy, to studying the tapes of the reunions of the six-year olds with their parents. Slowly, Cassidy found observable differences in how the children interacted with their returning parents. She showed that the internal working model had rooted in six-year olds. These studies moved child attachment theory closer to its roots in psychoanalysis.

17. They Are Leaning Out for Love: The Strategies and Defenses of Anxiously Attached Children, and the Possibilities for Change. Insecurely attached (ambivalent or avoidant) children suffer unconscious strategies for coping with their primary caregiver’s neglect or unavailability or indecision, and the furious emotions that well up within such a child, once slighted. Given the diversity among people, all generalizations neglect special coping skills of some, special alternative relationships, and mixtures and degrees of relating strategies. Nevertheless, some generalizations seem warranted. Ten percent of middle class children are ambivalently attached. They over-focus on mother, wheedling to get attention, finding dangers to be protected from. In the end, they fail to

believe they have what it takes to get what they want from another person. Klein emphasized the roiling emotional life of the insecurely attached child, the rage and aggression, self-loathing and anxiety. Ambivalent children fail to learn that mother can weather their rage while it dissipates, and suffer no ill-effects, while for the child, his rage seems uncontrolled and permanent. Ambivalent children often have unconscious fantasies of grisly retaliations against their love object. Refusal to receive offered love, when that is what is desperately wanted, is a form of aggression. It may relate to anorexia and sadistic emotional injury of others, all coupled with a fatalistic sense of repeating mother’s history. Twenty percent of middle class children are avoidantly attached. Such children turn off, seeming to need no love. The child preserves proximity to mother, without really relating to mother. He develops anger toward mother and others, grandiose self-deceptions, feelings of superiority, and detachment. His anger threatens all relationships. Avoidant and ambivalent attachment arises from unsuitable mothering, ignorance of the damage caused by absences, and premature training for independence, life stresses, overwhelmed parents, and parental psychological dysfunctions. A child with anxious maternal attachment may keep more secure patterns of relating alive if she is able to find an adult who relates to her in a secure manner. Still, the coping mechanisms of anxiously attached children tend to drive away those who might substitute. Where caregivers can persevere through hostility and ill-behavior, some insecurely attached children learn to attach to that caregiver and begin to take care of their intimacy needs. The child may change, but does so more readily when young. Adolescence and adulthood make all more committed to familiar patterns. A great problem in anxious attachment lies in the child’s ability to trigger aspects of the parents’ personalities that are themselves disturbed. Anxious attachment is frustrated love. Addressing anxious attachment requires consistent and persistent return to the child, in the face of ill-behavior and rejection, until the child builds up, as from little bricks, a revised interior model. Often, especially when the child’s anxious attachment derives from the parent, this proves difficult for the parent, because it flies in the face of the parent’s own internal working model. Even the most difficult avoidant child leaves open opportunities for changed relations with parents. But the opportunities are few, and likely to be initially rejected. Karen recounts his own, and another child patient’s experiences of anxious attachment, and moving away from it. A difficulty lies in the anxiously attached child’s interpretations of positive emotions. In his experience of mother, she remains outwardly warm, while inwardly demanding and unavailable. The child associates pleasant emotions with hidden coercion. Such a child exemplifies a fourth category of attachment: disorganized attachment. He exhibits characteristics of both avoidant and ambivalent attachment. Lieberman began to treat mother and child in therapy. As mothers of anxiously attached babies were able to recount their own sordid pasts and sufferings, the mothers’ ability to attach to their babies surged. After one year of such intervention, none among 100 anxiously-attached mother-child pairs could be distinguished from pairs with initial secure ratings. The Strange Situation test is a research tool, not a forensic measure. It should not be used to assess individual parent-infant dyads.

18. Ugly Needs, Ugly Me: Anxious Attachment and Shame. Consistent, reliable care builds self- esteem in children. They believe they deserve love and can affect their world positively. The anxiously attached child views himself negatively, feeling shame at what he is. The ugly distrust, bitterness, and hatred within he may feel the need to hide. The anxiously-attached child learns shame about those aspects of himself that his parents reject or ridicule. Where anxiously-attached children cope poorly with parental emotion and their responses to that emotion , shame emerges. The negative emotions of a securely-attached child get expressed, and parents respond to those communications meaningfully. Mothers of anxiously-attached children tend to ignore the negative emotions of their children. Parental anger, expressed volubly or suppressed, tends to make a child feel worthless. The parent of a securely-attached child expresses anger in a self- restrained manner, aimed to help the child process that anger. A parent whose dependency needs were unmet as a child finds it difficult to tolerate his child expressing that same unmet need. The child meets over-reaction or a dismissive rebuff. Shame may have evolutionary import. We evolved needing attachment; when that need is unmet, shame is attachment’s opposite.

Childhood attachment quality rings forward through personal histories. There are other influences (intellectual ability, non-maternal attachment figures, resilience, adult choices). But attachment style rumbles beneath all.

19. A New Generation of Critics: The Findings Contested. Jerome Kagan criticized all of attachment theory’s premises. Anxious attachments reflect, not parental style, but a child’s temperament and its learning of parental values. Human intimacy crumbles; we invent mythologies to support it, like maternal nurture in attachment theory. Socially-deprived infants do recover in appropriately nurturing environments, Kagan argued. Kagan’s line of thought went like this: Children differ. They change, as their environments change. Why focus on year one? The “internal working model” is an unlikely theory for which there exists little support. Temperament matters most, then environment. In response to Kagan’s assertions, Sroufe undertook analysis of children who had showed noteworthy declines or improvements from their early assessments. Sroufe found that shadows of initial attachment formats remain, even where there is overall improvement. And there are instances of marked decline from formerly secure attachment. Longitudinal studies by Crowell and Waters, show that among the earliest tested children, now adults, 69% retain their Strange Situation classifications, and when adjusted for untoward life events (death of mother), 77% retain those classifications. Karen concludes: a) attachment patterns exist, b) they show up as early as year one, and c) the affect people later in life, often deeply, becoming a component of our individual psychology. Having occurred pre- lingually, they are deeply buried and difficult to access. The attachment process continues past the first year. But the first year is critically important. All should want mothers and their babies to get off to a good start in those crucial early months. The Grossman longitudinal studies in Germany (Bielefeld) failed to replicate the American findings in very young children. Many more were avoidant (two-thirds insecure, and one half avoidant), which Grossman’s attributed to German cultural value for independence. Mothers pushed away infants as young as ten months, wanting them to begin self-restraint. A second study (Regensburg) closely matched the American results. Grossmans found that securely attached children profited from that orientation. The mothers of securely attached children were “less” German in their expectations, and better adapted personally. The northern German mothers were less rejecting than rejecting American mothers. There were more reversals of orientation in the German children. Michael Lamb’s critique of Ainsworth’s studies cut to their core, and his tone was insolent. Professional colleagues declined even to comment in response to his articles. Lamb’s critique, however caustic, made points: the Strange Situation studies addressed small numbers of individuals, there were few cross-checks, and replication studies were few. Karen finds these issues relevant and lingering.

Part IV. Give Parents a Break! Nature-Nurture Erupts Anew

20. Born That Way? Stella Chess and the Difficult Child. Advice on child-rearing mutated from the “do-not-coddle” position of Watson to permissiveness in the 1950s. Benjamin Spock noted pervasive confusion and problems raising children. Stella Chess noted that attachment theory made mothers feel encompassing guilt about their children’s problems. Chess believed that children have temperaments, and these interact with nurture to produce personality traits. David Levy too found that maternal style interacted with the underlying predilections of the child. Chess conducted a longitudinal study of temperament’s effect on later problems. She assessed 136 infants for activity level, the reliability of his or her biological rhythms, whether the child favored or disfavored new experiences, how the child adapted to new events over time, the intensity of reaction to discomforts, how intense new experiences have to be before eliciting response, general mood, distractability, and length of focus on an activity. Chess, from these characteristics, defined four types of temperament: difficult, slow to warm up, easy, and mixed. Some parental temperaments match poorly with their children’s temperaments, with expected poor outcomes. As Chess’s study progressed, some seventy percent of difficult temperament children suffered problems requiring therapeutic attention. In Chess’s view, temperaments interacted with environment to produce personality traits. Differing parental responses produced

markedly different outcomes for children of like temperament. Culture and the wider social circle also produced effects. Chess’s research, which continued for three decades into these children’s lives, emphasizes that one size does not fit all. Parents must adjust their approaches to the temperaments of their children. Chess’s colleagues disparaged the attachment theorists, and vice versa. Attachment theorists criticized Chess’s timing, reliance on parental self-reports, typology of temperaments and the underlying characteristics, and emphasis on fit between parent’s and child’s respective temperaments. The parties polarized, then warred.

21. Renaissance of Biological Determinism: The Temperament Debate. Both Bowlby and Ainsworth recognized infant temperament differences, but did not emphasize them. Genetic studies, however, showed massive hereditary influence on personality, as demonstrated by separated identical twins studies. These studies showed twin convergence on matters often deemed mere historical or cultural trivia. Twin studies had extensive influence. Physiological studies confirmed twin similarities on a deeper biological level. Kagan’s research on fearful toddlers led him to believe that people come in breeds, like dogs, and do not change their breed. Families have influences, but they are influencing the outcome of a genetic basis. Behavior geneticists attacked attachment theory in two ways. First, they assaulted the blame of mothers and the neglect of the idea of “poor fit” between mother’s and infant’s temperaments. Second, the behavior geneticists attacked the research underlying attachment theory, arguing that it in fact measured temperament, not attachment. Avoidant and ambivalent babies became the core dispute, each side arguing that the observed behaviors related to temperament or maternal attachment, depending on ideological commitments. Researchers turned to studies. Van Boom chose 100 babies classified highly irritable at birth. Fifty mother-baby pairs were given six hours of counseling about how to relate to a fussy baby. The other fifty received no such counseling. More than twice as many of the irritable babies whose mothers received counseling were classified as securely attached at one year. Clearly, maternal style was a critical, but not sole, determinant of secure attachment. Intense irritability in babies (temperament) also played a role. Behavior geneticists have found genetic influences, much to the detriment of environmental ones. Heralding genes and temperaments, they have neglected environmental influences, even when those are patent. Despite geneticist claims to the contrary, first relationships remain potent influences on the person’s ultimate ability to form lasting, meaningful relationships. Babies attune to their caregiving environment. What we have assumed to be temperamental, may in fact be environmental caregiving mismatch. The debate has been beneficial to both attachment theorists and behavior geneticists. Each is coming to grapple with the data of the other.

22. A Rage in the Nursery: The Infant Day-Care Wars. Attachment theory raised questions about the monumental social shift of women from homes to the workplace. Was mother work injuring infants? Was daycare imperiling future attachment? Non-attachment theories and feminists attacked the attachment findings. Research comparing daycare to home care was contradictory. The studies of daycare placements suffered from massive circumstantial differences and some poorly structured investigations. Bowlby asserted, in his blunt style, that mothers should be primary caregiver for their children and not work until the child was ready to go to pre-school (age three). This made Bowlby the feminist demon, until Belsky took over the job. After initially resisting the data, Belsky opined that more than twenty hours per week in daycare for a first year child created an increased risk for anxious attachment and its behavior problems. Feminist political operatives and their academic supporters scourged Belsky for questioning their orthodoxies. Regardless the political fallout, the data showed that the Strange Situation identified insecure attachment in just the sort of curve one would expect: descending insecure attachment as one progresses from abused children, to children of depressed mothers, to children spending more than twenty hours per week in daycare, to children from stable homes with less than twenty hours per week in daycare. Belsky’s predilection for blunt talk and seeking volatile venues (Good Morning America) partly explains his demonization. Belsky’s research, despite making him a pariah, brought the issue to the fore. Does daycare harm infants? Studies of daycare show dramatically low quality care (9% offer high quality care). In Vandell’s study of Dallas third- graders, children who as infants had thirty or more hours of daycare scored lowest in emotional

well-being, work habits, relationships with other children, and compliance with teachers. Bates’s study found a direct correlation between infant hours in daycare and problems with aggression at age five. European studies, where quality of care is high, show that such high quality care can not only match, but may outperform, parental care. Such high quality care is rare in the United States. But in the media, there is reticence to say so. And the business culture of the United States fails to accommodate the needs of parents without attendant loss of career possibilities. The attachment community of scholars seem to have reached consensus that parents can put infants in high-quality alternative care and not injure that baby. Still most developmental psychologists believe that a parent should stay home with his or her infant for a period of months following birth. Nanny care has the same problems as daycare in increasing insecure attachment. Government policy should make high quality day care available to all citizens. But there is a lack of political will in this regard.

23. Astonishing Attunements; The Unseen Emotional Life of Babies. Babies are born social. They recognize faces, prefer mother’s smells, and remember emotions. Mothers attune to their babies, giving them security in the midst of their vulnerabilities. An infant’s emotional health is erected upon the scaffold of his mother’s attuned involvement for an extended period of that child’s infancy. Mother-child relational exchanges came under examination. Researchers found an intricate skein of mother-child interactions in which the child led and the mother followed, but elaborated in variations of intensity, timing, and length. Mothers who synchronize well with their infants create secure attachment. Intrusive mothers foster avoidant babies. As language develops, the mother’s interaction serves to delimit the range of acceptable behavior and expression. Research shows that ambivalently attached children have mothers mis-attuned except as to baby’s fears. Mothers of avoidant babies mis-attuned to baby’s negative feelings, but gloried in his play. Well-attuned mothers lose themselves in their babies for a time, but slowly emerge from their monomania to broader life, just as the infant himself seeks greater autonomy. He comes to see others having lives independent from his own. Infants at this stage adopt transitional objects (teddy bear, doll, favorite blanket). No mother is perfect, nor should she be. Mothers should be “good-enough,” as Winnicott put it. The attachment patterns of a person tend to project into his or her own parenting.

Part V. The Legacy of Attachment in Adult Life

24. The Residue of Our Parents: Passing on Insecure Attachment. The parent-child dyad elicits powerful emotions, akin to new romance in intensity. Some of these emotions are negative: jealousy, hatred, or resentment, for example. In parents, when suppressed, these emotions later emerge veiled, recapitulating the parent’s childhood emotional experience of home. Mary Main studied transmission of attachment styles from parent to child in a Berkeley study. Main found that adults’ recollections of their attachment childhood fell into three categories, reflecting the Strange Situation categories: 1) Secure-autonomous (= secure attachment). These parents speak with ease and perspective about parents and childhood, and rely on important relationships. A great majority of their children are securely-attached. 2) Dismissive-of-attachment (= avoidant attachment). These parents speak haltingly of childhood and parents, and feel compelled to respond to the interviewer’s impertinent questions. Such parents have recollections that contradict their representations of their parents and childhoods. Three-fourths of the children of dismissive parents are avoidantly-attached. 3) Pre-occupied with early attachments (= ambivalent attachment). These parents are still actively engaged with their childhood parental lapses. They suffer little appreciation of their own role in failed and failing relationships. The great majority of their children are ambivalently- attached. Fonagy’s research with mothers before birth enabled him to predict with seventy-five percent accuracy the attachment style of the subsequently-born child in the Strange Situation. These mothers’ interviews showed that they were projecting, often unconsciously, their own childhood experiences upon the as-yet-unborn experience of their infants. The degree of

researcher subjectivity in rating adult subjects is substantially higher than among Strange Situation researchers. Fonagy and Main found that what differentiates secure from anxious adults is the ability to understand their own and others’ inner motivations. The avoidant mothers were distant, un- insightful about their own feelings, and failed to lead their infant into comfort with the child’s inner world, just as the parent remained uncomfortable. The avoidantly-attached mother keeps the baby at a distance, because proximity re-opens the mother’s own infancy, so filled with pain and anxiety. Mothers of anxiously-attached infants may perpetuate their child’s anxiety by being inconsistently available. This theory explains much. But it also fails to explain much. Why do some children have anxiously-attached parents but grow up secure? What role do fathers play in attachment types? Parents, in order to avoid reproducing their own insecurities and attachment defects in their children, need to face their own histories and its effects upon their own psychologies. And that is painful.

25. Attachment in Adulthood: The Secure Base vs. The Desperate Child Within. Karen tells the tale of an avoidant, but talented, judge who goes berserk after losing his adoring mistress, and ends up in jail. Bowlby believed that attachment style governed the emotional tenor not only of parenting, but of all of a person’s emotional life. Studies find that emotional health creates self- reliance conjoined with appropriate dependence on trusted others, and derives from a home with two loving, emotionally-available parents and a mother devoted to the child’s security in infancy. Solitude is healthy only when it one departs for a time from a secure social base, to which one returns after an appropriate period. Ambivalent or avoidant persons experience debilitating loneliness, and tend to grow addicted to people (for ambivalently-attached persons) or to work things, promotions, and obsessions (for avoidantly-attached persons). Loss of an attachment figure can debilitate even adults. Those who suffer such griefs slowly restructure their emotional lives to rely on new attachment figures. But avoidant persons fail to appreciate the damage losing important relationships will inflict upon them. Ambivalent persons grow panicked about separations and resist them. Hazan and Shaver studied self-reported attachment styles in adults. In romantic relationships, secure partners trust their lovers and accept their flaws. Avoidant partners fear intimacy, reject feelings of love, and suffer jealousy. Ambivalent partners obsess over their lovers. In adult exploration of life, secure persons explore and prosper. Ambivalent persons procrastinate, lose focus, and fuss over interpersonal matters. Avoidant persons work too much, letting tasks abort social relations, but prosper financially. But there are limitations to the Hazan and Shaver methodology. Self-reports may be inaccurate, and the romantic reports may reflect more the current love interests of the interviewees than their long term attachment styles. Adult attachment may be substantially more nuanced, with alcoves of different relating styles for different circumstances. Some researchers believe that the three categories are too simple and fail to comprehend mixed states and complex, changing relationships. Psychoanalysis has begun reintegrating attachment theory, revising the pre-oedipal states to accommodate the attachment research. There is some correlation between attachment states and psychiatric diagnoses. Nevertheless, attachment styles are not pathological in all circumstances, but rather broad, general observations of relational styles. Anxious attachment may, for some, become pathological, especially under severe stresses. Karen draws explicit links between severe anxious attachment and narcissism, compulsive personality, schizoid personality, and borderline personality. Karen appreciates Roger Kobak’s attachment continuum, which links normal experience of attachment issues to pathological states as degrees of experience.

26. Repetition and Change: Working Through Insecure Attachments. In transference (a psychoanalytic term) a patient projects childhood roles upon his therapist: parent, friend, lover. Attachment theorists find people transferring in many circumstances, especially when the emotional trigger is potent. Sullivan argued that we each have a self developed in response to some primary figure in our infancy, and we play out those various selves throughout life. Freud believed that we prefer relationships that recapitulate our early models, even when those relationships are unsatisfying. Freud called this “repetition compulsion.” Bewitched by failings

of parental attachment, the adult repetitively seeks the same sort of pain experienced as an infant. Worse, the person caught in such a pattern rejects better loves because they feel strange to him. Especially in sexual relationships, the ambivalent or avoidant adult plays out dramas of rejection, though secure relationship may be offered. Shame recurs. Ambivalent adults feel shame. Avoidant adults evade and suppress shame. Adults may find other models from childhood that assist them in coping with anxious attachments. Therapy may help, providing insight and safe conversation from which to begin to cope. The clean and simple world of the therapeutic relationship can provide a venue in which to experience self-sympathy and self-love, leading eventually to a reappraisal of one’s thwarted early attachment needs and self-reflection that is healing. In secure love relationships, people go over and over things that could not be addressed with parents, sometimes resolving persistent emotional tangles. The marriage becomes a secure attachment at the core of life. To break the intergenerational cycle of transmissible emotional damage of anxious attachment, one needs a secure attachment figure in childhood, or in-depth psychotherapy, or a stable and lovingly responsive spouse. With these helps, anxiously attached people may not engender anxious attachment in their children and associates. For some, having a baby offers opportunity to cope with anxious attachment issues. We may respond to parental inadequacies in two ways, neither of which is very helpful: first, we feel they are not to be criticized, since, despite their flaws, they provided for us, or, second, we rage against their flaws. Karen believes one must experience the injuries parents inflicted in one’s childhood, but move past them thoughtfully. Though childhood cannot be improved once its time has passed, one can adjust one’s attitudes now, re-examine both oneself and one’s parents from an adult perspective, and adequately mourn the losses one experienced as a child.

Part VI. The Odyssey of an Idea

27. Avoidant Society: Cultural Roots of Anxious Attachments. Pre-modern society consisted in community life: families, towns, guilds, extended friends and support, churches. Despite interpersonal bickering and absence of personal freedom or social mobility, people were secure and knew they were part of their community. Their many intimacies were attachments. In the modern era, industry and politics combined to erode that sense of place and safety. Competition replaced cooperation. The sphere of attachment shrank dramatically. People were not sure they had a place in society. Children lost their many associates, and were left with two (possibly, one) attachment figure: the parent. And that parent became unsupported by family and friends in many instances. People moved, dousing reciprocity (how can you help me if you will be across the country thirteen months from now? So, why should I help you now?). Life’s pace accelerated, leaving less social interaction. Every member of modern society became anxiously attached to that society. In primitive Yequana society (Venezuelan jungles), Leidoff, who lived with the tribe for more than two years, found a world of attachment. Babies were carried by mothers constantly, until they crawled. Children were securely-attached. Attachment affections persevered throughout adulthood. Leidoff concluded that modern society has lost much, including natural child-rearing processes. She urged recapture, with mothers staying home with babies, carrying them, instant response to distress, and sleeping with infants and young children. Leidoff’s study, and others like it, instill a sense of loss when reflecting on modern society. Many are attempting to re-introduce some of the Yequana child-rearing practices into their modern lives. Karen advocates introducing greater simple helping and support structures in our society. In longitudinal studies, the children of poor families provided such supports show far fewer dysfunctions ten years out. Poverty itself does not generate anxious attachment. But early neglect in the parents’ own lives and scant support structures dramatically increase the likelihood of anxious attachment, even learning disabilities. Experts recommend paid family leave, strong day care standards and financial supports, comprehensive public health care, income support for poverty-stricken families, and appropriate therapeutic interventions. Age isolation should be met with giving children regular opportunities to relate to younger and older persons. And teachers should be trained not to worsen insecurely attached children by meting inappropriate discipline. All these programs would be very expensive, and really cannot provide what can only come from caring individuals. We need structures that support and improve the chances of intimate, caring

human relationships. Nevertheless, our current pattern of moral condemnation, incarceration, and disgust fail to address the underlying problem. Investing in attachment, society will generate fewer problematic personalities at a cost, but will reap the benefits of having better overall life conditions for most.

28. Looking Back: Bowlby and Ainsworth. Late in life, Bowlby wrote a psycho-biography of Charles Darwin. He painted Darwin as a study in spoiled attachment, having lost his mother at age eight (and hence sent to boarding school) and tormented by a distant father. Darwin suffered debilitating anxiety his entire life. Like Darwin, Bowlby loved his friends. But Bowlby did not want intimacy with them. Bowlby was avoidant. He was gratified that some of his concepts won the intellectual day. The analytic community has come to accept Bowlby; some even embrace him. Ainsworth, like Bowlby, had avoidant features in her personality. She and Bowlby were friends and admired one another’s work. Ainsworth had a keen mind, and wrote well. Ainsworth’s research worked a revolution, challenging the fervent desire among feminists that both parents be able to work without damage to children. Unlike Bowlby, Ainsworth was shy. She just loved the details of her work. Both Ainsworth and Bowlby failed to include some features of infant life in their theory: fantasy, sexuality, aggression. Fathers were neglected. Later adolescent adaptations did not command attention. So too, attachment to non-parental others. How does attachment affect character? Does insecure attachment lead to neurosis? What other systems are developing in children as they age, and how do those relate to attachment? All classifications fail to encompass the complexity of individual life. Attachment styles are no exception. Both Bowlby and Ainsworth sought reform, political and social, regarding children and the way we think of them. Loving attachment to a reliable caregiver is a primary motivation. This fact flies in the face of much in our culture. We are distracted. Still, our need for closeness, affection, touching, understanding, our desire to work through losses: these are biological demands. We cannot avoid them by denying our feelings about them.

Appendix: Typical Patterns of Secure and Anxious Attachment. Karen charts short descriptions of secure, ambivalent, and avoidant attachment at various developmental phases through individual life.